The proposed research is to develop the analysis of tubular structures into a standard electron crystallographic approach that can be used to (1) solve the structure of the nicotinic acetylcholine receptor at atomic resolution and (2) solve the three-dimensional structure from any tubular specimen in a semi-automated approach. To accomplish aim 1, improvements in the image processing routines will be made to extend resolution beyond the current 4.6A. Areas include more accurate measurement and correction of distortions, refined treatment to account for noise from overlapping Bessel terms, and corrections for Ewald sphere curvature, beam tilt and the contrast transfer function. Aim 2 will be achieved by analyzing tubular crystals of proteins being studied by other investigators using the methods established for the acetylcholine receptor after they have been automated. The rigorousness of the image processing and the broad applicability of the automated methodology will be determined.